Spatial Cluster Analysis of P. vivax and P.malariae Exposure Among Haitian School Children Sampled Between 2014 And 2016 Open Access

Oviedo, Adan (Spring 2018)

Permanent URL: https://etd.library.emory.edu/concern/etds/cn69m418x?locale=en
Published

Abstract

 

Malaria is a parasitic disease associated with flu-like symptoms. Haiti’s failure to eradicate malaria during the 1955-1969 global intervention effort testifies to the need for accurate surveillance methods in sub-microscopic low-transmission settings. Protocol within the Transmission Assessment Survey (TAS) employs multiplex bead technology to determine exposure to Plasmodium Merozite Surface Protein 1 (MSP1). The goal of the project is to describe the distribution of P. malariae and P. vivax exposure as well as identify significant hot-spot foci in Haiti. Access to the de-identified TAS was granted by the CDC. Environmental variables were attained from open source sites and analyzed using univariate logistic regressions. K-functions were calculated using a maximum radius of 50km with 2.5km intervals and confidence intervals using 999 permutations. Moran’s I was calculated using inverse distance weights, 999 permutations, and a threshold distance of 20km. Kulldorff’s Scan Statistic was calculated using discrete purely spatial Poisson modeling and 999 Monte Carlo simulations. 21,670 children were tested for P. malariae in 679 schools. 24,510 children were tested for P. vivax in 787 schools. 275 (1.27%) and 113 (0.46%) children were determined seropositive for P. malariae and P. vivax, respectively. K-functions suggested school aggregation at all intervals up to 50 km. Weighted K-functions in north Haiti suggested aggregation at limited intervals. Weighted k-functions in south Haiti suggested spatial randomness at all intervals. Moran’s I in the north suggested autocorrelation for P. malariae (I=0.153, p<0.01) but not for P. vivax (I=-0.016, p>0.05). Moran’s I in the south suggests no autocorrelation for P. malariae (I=-0.001, p>0.05) or P. vivax (I=0.062, p<0.10). Kulldorff’s Spatial Scan suggested five hot-spots for non-falciparum malaria exposure. No environmental variables were associated with seropositivity. Cluster A (LR=20.8, p<0.001) was determined a significant hot-spot of interest. Due to limitations, clusters B (12.3, p<0.01) and E (LR=10.1, p<0.05) are potentially of interest but need further review. Clusters C (LR=22.1, p<0.001) and D (10.4, p<0.05) likely result directly from limitations in methods. This study serves as a baseline distribution of non-falciparum malaria within Haiti’s borders intended to provide suggestions for focused and thorough active surveillance projects conducted in the near future.

 

Table of Contents

 

Background………………………………………………………………………………………...1

 

Materials and Methods…………………………………………………………………….….…11

 

Results…………………………………………………………………………………………….15

 

Discussion………………………………………………………………………………………...19

 

References………………………………………………………………………………………...26

 

Tables and Figures………………………………………………………………………………..33

 

Appendix A……………………………………………………………………………...………..39

 

About this Master's Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
School
Department
Degree
Submission
Language
  • English
Research field
Keyword
Committee Chair / Thesis Advisor
Committee Members
Partnering Agencies
Last modified

Primary PDF

Supplemental Files